Effect of Airway Blood Flow on Airflow 1 , 2 ALAIN LOCKHART, A. TUAN DINH·XUAN, JACQUES REGNARD, LAURE CABANES, and REGIS MATRAN
Introduction As early as 1935, De Burgh Daly (1)proposed that swelling of the bronchial mucosa played a part in the causation of an asthmatic attack and that "the reactions of the bronchomotor ... apparatus of the lungs depend upon the integrity of the bronchial vascular system." In 1943, through use of rigid bronchoscopes, obstructive congestion and edema of intrapulmonary airways were documented in patients with left heart failure (2), and in 1950, they weredocumented in asthmatic subjects after inhaled allergen challenge (3). At about the same time, results of physiologic measurements before and after the subcutaneous administration of epinephrine indicated "... the likelihood that adrenaline produces pulmonary vascular constriction rather than bronchiolar dilatation in certain cases" (4). In 1960, dilatation of the capillary blood vessels beneath the basement membrane and associated mucosal edema were described as "a most striking feature" of the bronchi of subjects who had died in status asthmaticus (5). In 1961, the role of airway narrowing caused by congestion and edema in occurrence of exertional dyspnea and cardiac asthma in congestiveheart failure was again emphasized (6). Thus, the case for v'ascular factors being major contributors to airway obstruction in both left heart failure and asthma was already very strong in the late fifties and early sixties. It is, therefore, surprising that it fellinto oblivion for about 20 yr and that research interest in this area was rekindled only during the last decade. This review will purposely focus on subglottic airways since effects of nasal congestion and vasoconstriction, respectively causing obstruction and increased patency of nasal airways through effects on blood content of venous sinusoids, are fully documented (7). Because there is no available technique to measure tracheobronchial blood flow and its distribution in humans, direct evidencethat airway blood flow influences airflow has not been obtained in humans and was only accrued from studies in experimental animals. In the first part of this report, we review recent experimental work in which the investigators provided simultaneous measures of airway blood flow and lung mechanics. Next, we examine possible mechanisms whereby changes in local blood flow might affect lung mechanics. Lastly, we review human studies in which the investigators attempted to provide evidence of the influence of airway blood flow on airway patency. Studies in Animals Concomitant Bronchial Obstruction and Vasodilatation In intact sheep anesthetized with pentobarbital sodium, inhaled histamine caused an in-
SUMMARY Resistance to gas flow of an airway is a function of both airway smooth muscle tone and thickness of the airway wall internal to the outer ring of airway smooth muscle. Schematically, the increase in airway resistance caused by shortening of airway smooth muscle may be potentiated by a concomitant increase in airway wall thickness caused by vasodilatation of the bronchial vessels and/or microvascular leakage. Conversely, bronchial vasoconstriction may limit to some extent the increase In resistance to gas flow caused by airway smooth muscle shortening and/or congestion and edema of the airway wall. Many endogenous paracrine mediators, putatively Involved in asthma and bronchial hyperresponsiveness, have both bronchomotor and vascular effects. The overall effects on resistance to airflow of endogenous or exogenous agents depend not only upon pre-existing airway smooth muscle tone and pre-existing condition of bronchial vessels but also upon two factors that facilitate microvascular leakage, namely, inflammation of the airway wall and outflow pressure of the bronchial circulation, which is close to left atrial pressure. AM REV RESPIR DIS 1992; 146:819-S23
creasein both bronchial blood flow (measured by an electromagnetic flow probe) and lung resistance with a similar time course, reaching a maximal effect within a fewminutes and progressivelyreturning toward baseline values within 1 h. Bronchial obstruction" and vasodilatation of the bronchial circulation were prevented by the histamine H, antagonist, chlorpheniramine, and the histamine Hz antagonist, metiamide, respectively (8). In pigs anesthetized with pentobarbital, bronchial blood flow was measured with an ultrasonic flow probe while lung resistance and compliance were also recorded. Histamine aerosol caused both bronchial obstruction that was almost completely blocked by the histamine H, antagonist, terfenadine, and vasodilatation that was reduced by only 55 to 60070 by the combination of terfenadine and the histamine Hz antagonist, cimetidine (9). Vagal stimulation also caused both bronchial obstruction and vasodilatation (10). In the same preparation, acetylcholine, nebulized PAF-acether, and prostaglandin D, induced both bronchial obstruction and vasodilatation (9). Vagal-stimulation-induced bronchial obstruction was prevented by atropine, whereas the vasodilatation was resistant to atropine as it was likely due to activation of capsaicinsensitive sensory nerve endings (10). In both sheep and pigs sensitized with Ascaris suum, the early response to nebulized antigens was characterized by a marked and progressive increase in both lung resistance and bronchial blood flow, which peaked within about 10 min and subsided after 90 min. These results were best explained by a histamine-H, -mediated bronchoconstriction and by vasodilatation induced by histamine and cyclooxygenase products. The latter was probably due at least partially to activation of capsaicin-sensitive sensory nerves (9). It was, therefore, hypothesized that changes in bronchial blood flow during the acute response are a sensitive index of mediator re-
lease (11). About half of the sheep had both early and late responses as assessedby changes in lung resistance. Interestingly, the decrease in bronchial vascular resistance preceded the late-phase bronchial obstruction by 60 to 90 min (12). In those experiments, vasodilatation and bronchial obstruction are not necessarily causally linked since both may result from release of mediators by inflammatory cells invading the bronchial wall (12).
Dissociation of Bronchial Obstruction and Vasodilatation Different stimuli caused changes in the bronchial circulation ofpentobarbital-anesthetized pigs with no concomitant changes in lung resistance and dynamic compliance (10). First, inhaled capsaicin in the presence of autonomic blocking agents caused a marked increase in bronchial blood flow but no change in pulmonary mechanics (10). Prior systemic administration of capsaicin significantly reduced capsaicin-induced vasodilatation (10). Furthermore, vasodilatation, which was likely due to release of sensory neuropeptides, was
1 From the Laboratoire de Physiologie Respiratoire, Faculte de Medecine Cochin Port-Royal et Laboratoire d'Explorations Fonctionnelles, Hopital Cochin, Paris, France. 2 Supported by Grant 885012 from INSERM, by GRECO GDRS 15from Centre National de la Recherche Scientifique, Grant 2403Rll from Direction de la Recherche et des Etudes Doctorales, and by Laboratoires Synthelabo. 3 Correspondence and requests for reprints should be addressed to Alain Lockhart, M.D., Laboratoire d'Explorations Fonctionnelles, Hopital Cochin, 27 rue du Faubourg Saint Jacques, 75014 Paris, France. 4 "Bronchial obstruction" is purposely used throughout this report in place of "bronchoconstriction" because the former does not imply a given mechanism, whereas the latter suggests airway smooth muscle contraction alone.
519
520
LOCKHART, DINH-XUAN, REGNARD, CABANES, AND MATRAN
a prominent effect of substance P and/or of calcitonin gene-related peptide in the pig (13). Conversely, the major effect of substance P in the guinea pig is bronchoconstriction which probably accounts for both capsaicin- and ether inhalation-induced bronchial obstruction in this species (14). Second, atropine blocked the vagally induced increase in lung resistance despite the presence of an intact vasodilator response. Similarly, cigarette smoke, calcitonin generelated peptide, bradykinin, and prostacyclin caused bronchial vasodilatation but no change in bronchomotor tone (9). Lastly, on the one hand, the potent bronchial vasodilators prostaglandins E, and E, not only did not reduce airways patency but instead caused bronchodilation (9). On the other hand, vasoconstrictor agents, e.g., a-adrenergic agonists and neuropeptide Y, which increased bronchial vascular resistance, had little if any bronchomotor effect (10). Tracheobronchial blood flow increases with exercisein ponies (15),and with passive hyperventilation (16) and thermal panting in dogs (17). Although they were not measured, one could reasonably assume that lung mechanics were not affected by the increased blood flow in these experiments.
Interplay of Airway Blood Flow and Airflow Summarizing all these results, one is led to the conclusion that in some studies there is a concomitant increase of hindrance to airflow in subglottic airways and in bronchial blood flow, whereas in other experiments, changes in the bronchial circulation are dissociated from changes of lung mechanics in vivo. One could wonder whether it would be possible to reconcile these discrepant findings. Interpretation of simultaneous changes in bronchial blood flow and airflow caused by agents capable of a direct action on bronchial vessels and bronchial smooth muscle is apparently straightforward. For instance, histamine and acetylcholineare potent vasodilators of the tracheobronchial circulation (18), and they contract airway smooth muscle in vitro. It is not surprising, therefore, that they cause a concomitant rise in blood flow and resistance to airflow. However, the relative contribution of shortening of airway smooth muscle and vasodilatation-induced thickening of the airway wall (18) is unknown in vivo. Another apparently simple situation prevails when vasodilatation is associated with bronchodilatation, for example, when the pharmacologic agents used, e.g., prostaglandins E, and E z , are known to relax both vascular and airway smooth muscle. In such cases, effects on the internal diameter of the airways of vasodilatation-induced thickening of the wall is opposed by the effects of bronchial smooth muscle relaxation. However, simple mechanistic explanations cannot account for all experimental data. Indeed, increased bronchial blood flow can be associated with an almost unchanged hindrance to airflow although the agent used is
capable of relaxing (e.g., prostacyclin) or contracting (e.g., tachykinins) airway smooth muscle in vitro. Indeed, several confounding factors may modify the apparent interplay of changes of bronchial blood flow and airflow (figure 1). First, increase in wall thickness of pumpperfused dog trachea varies in an unpredictable manner according to the locally infused vasodilator used, i.e., there was no obvious relation between wall thickness and degree of vasodilatation or known capacity of the agent used for causing microvascular leakage (18). Second, measures of bronchial blood flow with a probe placed around the main bronchial artery tells nothing about blood flow distribution both longitudinally and across the airway wall (19).In the larger bronchi, the subepithelial vascular plexus is connected in parallel to the adventitial plexus by perforating vessels and is drained by bronchial veins. More peripherally, the two plexi merge in a single microvascular network that anastomoses to the alveolar capillaries and drains into pulmonary veins and left atrium (20). In unanesthetized sheep, histamine caused a twofold increase in total systemic blood flow to the airways (19). However, blood flow to the trachea and lobar bronchi increased three and six times over baseline, respectively. Blood flow selectively increased in the mucosal and submucosal layers. The histamine-induced increase in blood flow was to to 15 times over
AirwaywaJJ ttuckness
Endothelium nqhtness Pmv . Pores - Inftarnmation
Clearance from airways walls
Migration of blood cells Activation of mediators eg kinins
Airway smooth muscle tone - direct reflex
Fig. 1. A nonproportional Venn diagram of the interplay of airway smooth muscle tone (bottom circle), tracheobronchial blood flow (Otb) (left upper circle), and endothelium tightness (right upper circle) in the setting of subglotlic airway resistance (Raw). Airway wall thickness is influenced by both vasomotor changes of the tracheobronchial circulation and endothelium tightness. The latter also depends on microvascular pressure and structural changes affecting endothelial cells such as formation of pores caused by pharmacologic agents (e.g., histamine) or local inflammation. Increase in endothelial permeability is associated with diapedesis of inflammatory cells and plasma exudation, which in turn lead to activation of inflammatory peptides (e.g., kinins) and complement moieties. In addition, tracheobronchial blood flow affects clearance of vasoactive and branchoactive substances from the airway wall, thereby modifying magnitude and duration of their effects on the bronchi. Reflexly or directly caused changes in airway smooth muscle tone are major determinants of Raw. Effects on airway caliber of changes in length of airway smooth muscle depend on both site of affected airways and preexisting or concomitant wall thickness. Details in text.
baseline in more peripheral airways (1 to 5 mm in diameter) and five to 15 times over baseline in smaller airways « I mm in diameter). The increase in blood flow was accompanied by wall edema as evidenced by the increase in water content. In these experiments, there was little change in pulmonary blood flow and almost no interstitial or alveolar edema (19).In anesthetized sheep, isocapnic ventilation of dry gas increased blood flow to-fold in the proximal but not in the distal trachea even though mucosal cooling also occurred distally (21). Indeed, factors controlling mucosal blood flow are poorly understood both in large and even more in peripheral airways. In the sheep, tracheal mucosal blood flow lacks cholinergic muscarinic responsiveness. By contrast, tracheal mucosal blood flow shows vasoconstrictor and vasodilator responses to a- and ~-adrenergic agonists that are qualitatively similar to those for total tracheal blood flow (22). Third, effects of a given pharmacologic agent on airway microvascular permeability and, therefore, on airway wall thickness, vary with local conditions. Normal bronchial vessels are not permeable to particulate tracers (23,24). Inflammatory mediators, e.g., histamine, serotonin, sulfidopeptide leukotrienes, and PAF-acether, increase the permeability of the microcirculation in the airways. The increase in permeability is restricted to postcapillary venules and results from active changes in the configuration of the endothelial cells and the opening of gaps between these cells (23). Similarly, both antidromic stimulation of the vagal nerves in atropinized animals, and irritants that stimulate sensory nerve endings cause microvascu1arleakage from postcapillary venulesthrough release of substance P (23-25). Neurogenic microvascular leakage in the airways of experimental animals can be potentiated by mediators, e.g., prostaglandin E" which alone do not increase microvascular permeability in the trachea (26). Naturally occurring respiratory tract infections also increase neurogenic microvascular leakage in the rat airways, as compared with those of specific pathogen-free animals (27). Conversely, administration of a vasoconstrictor agent reduces pharmacologically induced plasma extravasation in the airways (28-30), as well as serotonin-, histamine-, and acetylcholineinduced bronchial obstruction in guinea pigs (31)and nitroglycerin-induced increase in peripheral airway resistance (32). This preventive action is likely due to the reduced microvascular pressure caused by the increase in arteriolar resistance and decreased blood flow in the tracheobronchial circulation. Fourth, congestion, and even more, exudation, caused by either an increase in intravascular pressure (32) or a pharmacologic vasodilatation (18), are capable of increasing airway wall thickness from the trachea down to the small peripheral airways. The effects of wall thickening depend not only on the magnitude of thickening per se but also on the size and location of the affected airways.
521
AIRWAY CIRCULATION AND AIRFLOW
In the trachea and proximal bronchi with a large cross section in relation to the mucosa, it is commonly accepted that mucosal thickening is unlikely to significantly narrow the airway lumen (33,34). However, this contention should be questioned. First, although it certainly applies to large animals, e.g., dogs and sheep, this probably does not hold true for smaller laboratory mammals whose main bronchi range from 1to 2 mm. In smaller cartilaginous and membranous bronchi, the microvascular network is located within the wall, which is made of circular smooth muscle embedded in connective tissue. These are the bronchi where airway wall thickening is most likely to encroach upon the lumen (31), the more so when shortening of airway smooth muscle reduces the external diameter. Indeed, it has been shown that congestion of bronchial vesselsincreased peripheral airway resistanceto gas flow (33).In the smaller bronchioles, which have thinner walls and are surrounded by alveoli, the increased wall thickness occurs at the expense of surrounding air spaces, with little reduction in airway lumen (34). Second, where marked edema is associated with vasodilatation, obstruction of major and second-order bronchi (3) and of sub segmental airways (35) has been documented at bronchoscopy at the time of the early response to allergen challenge in humans. With respect to effects of airway vascular engorgement caused by fluid infusion in animals, it should be emphasized that the concomitant increase ill airway resistance to gas flow results at least partially from reflexly increased vagal tone (36, 37). Therefore, the actual relationship of airway blood flow to airway diameter may well be modified by the pentobarbital anesthesia used in most studies. Lastly, changes in airway blood flow may modify responses to bronchial challenges in vivo..In dogs, the time constant of recovery from histamine-induced increase in peripheral airflow resistance was prolonged with interruption of bronchial blood flow (38). In sensitizedsheep, vasopressinprevented the rise in tracheal blood flow and prolonged the decrease in tracheal volume caused by inhaled Ascaris suum. Nitroglycerin infusion increased tracheal blood flow, and the subsequent administration of Ascaris suum caused neither a further increase in tracheal blood flow nor contraction of the trachea (39).Thus, vasoconstriction likely potentiates the effects of both exogenous and locally released bronchoconstrictor substances through reduction of their vascular clearance from the airway wall. Conversely, increased blood flow may speed up the elimination of spasmogenic substances from the airways, thereby blunting their bronchoconstrictor effects in vivo. In summary, effects of changes in blood flow to the airways on resistance to airflow are not accounted for by changes in airway wall thickness alone. Longitudinal and crosssectional distribution of blood flow, activation of sensory nerve endings, structural changes in the airway wall, modified clearance of bronchoactive substances are con-
founding factors superimposed on the geometric factor. Their effects on airflow may be either reinforced or blunted in experimental animals, much depending on the species studied and the underlying mechanism that predominates. Results of human studies are even more difficult to interpret since no direct measurements of airway blood flow are available. Airway Blood Flow and Airflow in Humans Evidence for effects of changes in blood flow on airway patency is at best circumstantial in humans since none of the available techniques measuring airway blood flow can satisfy both the criteria of accuracy and that of being noninvasive. It was hoped that laserDoppler flowmetry during fiberoptic bronchoscopy might provide direct information on mucosal blood flow in humans. Unfortunately, experiments in live and in dead sheep suggest that in its present form laser-Doppler flowmetry cannot be relied upon to measure airway wall blood flow during bronchoscopy because of background noise at zero flow and large site-to-site variability (40).
chial obstruction in response to u-agonists (48,49), we suggested that the protective effect of methoxamine was likely due to the preventive effects of methoxamine-induced hyperemia and microvascular leakage (47) as opposed to its disproportionate contractile effect on airway smooth musclein unchallenged asthmatics. Indeed, the bronchial response to an agonist may depend on the balance between its vascular effects and its effects on airwaysmooth muscle.Thus, prostacyclinmay cause either bronchodilation or bronchial obstruction in asthmatic subjects (50). Vasodilatation accompanied by plasma extravasation likely occurs spontaneously and during bronchial challenges other than exerciseor hyperventilation in asthmatic subjects. First, inflammatory lesions are found in biopsy specimens of bronchi from asthmatic subjects between asthmatic attacks and even more in symptomatic asthma (51), and vasodilatation and microvascular leakage are common features of inflammation (52). Second, many mediators putatively involvedin asthma and many substances used to demonstrate bronchial hyperresponsiveness in asthmatic subjects, e.g., PAF-acether, histamine, bradykinin, tachykinins, sulfidopeptide leukotrienes, and acetylcholine, with the exception of a-adrenergic agonists and f}-blockers (53, 54), cause vasodilatation and/or plasma exudation. It is likely their effects on airways microvasculature are potentiated in asthmatic subjects because of the presence of inflammatory changes in the airways, as has been demonstrated in the rat (23).
Studies in Asthma Local instillation of antigen under direct vision results in immediate blanching of the mucosa followed within minutes by mucosal edema. The latter is accompanied by hyperemia, so that the mucosa becomes increasingly red, reflecting the congested blood vesselsimmediately beneath the surface of the mucosa (35, 41). Studies in Left Heart Failure Exerciseand hyperventilation of dry and/or cold air are followed by an asthmatic attack Bronchial hyperresponsiveness to the bronin a majority of asthmatic patients. The choconstrictor and vasodilator agents, histapathogenesis of exercise-induced(hyperventi- mine and methacholine, is common in mitral lation-induced) asthma is not fully under- valvedisease and left ventricular failure (55). stood. The thermohydraulic demands on re- Bronchial hyperresponsiveness in such pagional heat and water exchangecertainly result tients is likely due to potentiation of the in an early increase in the tracheobronchial vasodilator properties of histamine and blood flow, almost from the start of exercise methacholine by the high pulmonary and traor hyperventilation. However, bronchial ob- cheobronchial microvascularpressure of these struction is delayed, being maximal only patients. Reasons are as follows. First, there about 5 to 10 min after exercise. There are is a significant correlation between the detwo possible explanations for the delayed gree of bronchial hyperresponsiveness and bronchial response. First, methacholine- pulmonary wedge pressure in patients with induced bronchoconstriction is prevented or mitral valvedisease(55).Second, pretreatment reversed by exerciseand hyperventilation (42, with the inhaled vasoconstrictor agent, 43), which suggests they exert a protective role methoxamine, fully prevented bronchial against nonspecific bronchial hyperreactivity hyperresponsiveness to methacholine in pathrough an unknown mechanism. Second, re- tients with left ventricular failure (56). Third, bound hyperemia probably occurs only af- in those patients, inhaled 132-agonist only ter, and not during, exercise as evidenced by caused partial reversion of methacholinea faster rewarming of the airway wall during induced bronchial obstruction, thus suggestearly recovery in asthmatic compared with ing that the latter could not be attributed to that in normal subjects (44, 45). There is, how- contraction of airway smooth muscle alone ever, no evidence suggesting the release of (56). Clinical implications of bronchial hyperpreformed, or newly formed, inflammatory responsiveness in left ventricular failure and mediators in bronchoalveolar lavage fluid in mitral valve disease are not clear to date. We exercise-induced asthma (46). In addition we have shown that exertional dyspnea is more have shown that pretreatment of inhaled severein patients with left ventricular failure methoxamine, a potent a-adrenergic agonist, and bronchial hyperresponsiveness than in prevents exercise-induced asthma in some those without a response to methacholine asthmatic subjects (47). Because asthmatic (Costes, Cabanes, Lockhart, et al.; personal subjects are exquisitelyprone to develop bron- communication). Furthermore, pretreatment
522
LOCKHART, DINH-XUAN, REGNARD, CABANES, AND MATRAN
Histamine Allergen Bradykinin LTc/D/E4
+
I B2-agonists
i
PGI2
1 1 IRaw\ IIRaw\
Acetylcholine
or 111
1
Ir-R....;..aw-~,....,I
of inhaled methoxamine mar kedly improved exercisecapacity in patients with left ventricular failure in Class III for dyspnea of the New York Heart Association (57). It is therefore likely that exercise-induced vasodilatation facilitates, in the presence of an elevated pulmonary venous (left ventricular enddiastolic) pressure, microvascular leakage in the airway wall, thereby causing some degree of bronchial obstruction in these patients. Consistent with this hypothesis is our finding that inflation of antishock trousers in the erect posture causes a mild degree of bronchial hyperresponsiveness to methacholine in normal subjects (58).
Fig. 2. Airway resistance (Raw) increases when rise in tracheobronchial blood flow (atb) is accompanied by contraction of airway smooth muscle (SM) with (left) or without (right) concomitant increase in permeability (P) of endothelium. Effects of increase in atb on airway resistance when airway smooth muscle tone varies according to balance between vasodilatation and bronchodilatat ion (middle) and to associated changes in microvascular permeability. Same format as figure 1 (t, +, NC = increase, decrease, no change, respectively; LTCIDIE, = leukotrienes C" D" and E,; PGI, = prostacyclin). Details in text.
two situations: (1) when thickening ofthe airway wall is abnormally important because of an elevated microvascular pressure in the airway microcirculation or because of the potentiating effects of inflammatory changes in the airway wall, and (2) when there is concomitant contraction of airway smooth muscle. Morphometric studies of the airways of experimental animals, similar to those carried out to document the role of airway wall thickening in human asthma (59), have not been done to our knowledge. Such studies are probably needed to further our understanding of the effects of changes in blood flow on airflow.
Concluding Remarks
References
Effects of changes in airway blood flow on airflow are far from simple (figures 2 and 3). Indeed, vasodilatation and plasma exudation cause thickening of the airway mucosa. Studies in animals, and circumstantial evidence in humans, suggest that this geometric factor is an important determinant of spontaneous or induced obstruction to airflow in
1. De Burgh Daly I. Interference of intrinsic pulmonary mechanisms as a potential cause of asthma. Edinburgh Med J 1935; 43:139-42. 2. Renault P, Paley PY, Lenegre J, Carouso H. Les alterations bronchiques des cardiaques. J Fr Med Chir Thorac 1943; 3:141-59. 3. Vallery-Radot P, Halpern BN, Dubois de Montreynaud JM, Pean V. Les bronches au cours de la crise d'asthrne. Etude experimentale, bronchoscopique et anatomopathologique. Presse Med 1950; 58:661-4. 4. Shedon MB, Otis AB. Effect of adrenaline on resistance to gas flow in the respiratory tract and on the vital capacity of normal and asthmatic subjects. J Appl Physiol 1951; 9:513-8. 5. Dunnill MS. The pathology of asthma, with special reference to changes in the bronchial mucosa. J Clin Pathol 1960; 13:27-33. 6. Rushmer FJ. Cardiovascular dynamics. 2nd ed. Philadelphia: Saunders, 1961; 470-1. 7. Cole P. Upper respiratory airflow. In: Proctor DF, Andersen 18, eds. The nose: upper airway, physiology and the atmospheric environment. Amsterdam: Elsevier, 1982; 163-90. 8. Long WM, Sprung CL, El Fawal H, et a/. Effects of histamine on bronchial artery blood flow and bronchomotor tone. J Appl Physiol 1985; 59:254-61. 9. Alving K. Airways vasodilatation in the immediate allergic reaction: involvement of inflammatory mediators and sensory nerves. Acta Physiol Scand 1991; 141(Suppl 597:1-64). 10. Matran R. Neural control of lower airway vasculature: involvement of classical transmitters and neuropeptides, Acta Physiol Scand 1991; 142(Suppl 601:1-54). 11. Long WM, Yerger LD, Martinez H, et al. Modification of bronchial blood flow during allergic airway responses. J Appl Physiol 1988;
Alpha-agonists in asthma
Alpha-agonists in LV failure
Fig. 3. A diminished tracheobronchial blood flow can be associated with an increased or decreased airway resistance depending upon magnitude of concomitant changes in both airway smooth muscle (SM) tone and endothelial permeability (P). Inhaled alpha-adrenergic (left) agonists cause moderate to marked bronchial obstruction in most cases of asthma, but they reduce it in the remaining cases, especially when bronchial edema is predominant. These agents invariably reduce methacholine-induced bronchial obstruction in patients with left ventricular failure (right). Details in text. Same format and abbreviations as figures 1 and 2.
65:272-82. 12. Long WM, Yerger LD, Abraham WM, Lobel C. Late-phase bronchial vascular responses in allergic sheep. J Appl Physiol 1990; 69:584-90. 13. Alving K, Matran R, Lacroix JS, Lundberg JM. Capsaicin and histamine antagonist-sensitive mechanisms in the immediate allergic reaction of pig airways. Acta Physiol Scand 1990; 138:49-60. 14. Martling CR, Matran R, Alving K, Hokfelt T, Lundberg JM. Innervation of lower airways and neuropeptide effects on bronchial and vascular tone in the pig. Cell Tissue Res 1990; 260:223-33. 15. Manohar M. Tracheobronchial perfusion during exercise in ponies. J Appl Physiol 1990; 68:2182-5. 16. Baile EM, Godden OJ, Pare PD. Mechanism for increase in tracheobronchial blood flow induced by hyperventilation of dry air in dogs. J Appl Physiol 1990; 68:105-12. 17. Baile EM, Guillemi S, Pare PD. Tracheobronchial and upper airway blood flow in dogs during thermally induced panting. J Appl Physiol 1987; 63:2240-6. 18. Laitinen LA, Robinson NP, Laitinen A, Widdicombe J G. Relationship between tracheal mucosal thickness and vascular resistance in dogs. J Appl Physiol 1986; 61:2186-93. 19. Kramer GG, Lindsey C, Wu CH, Mertens S, Russell LA, Cross CEo Airway blood flow distribution and lung edema after histamine infusion in awake sheep. J Appl Physiol1988; 65:1847-54. 20. Deffenbach ME, Charan NB, Lakshminarayan S, Butler J. The bronchial circulation: small, but a vital attribute of the lung. Am Rev Respir Dis 1987; 135:463-81. 21. White DA, Parsons GH. Tracheal blood flow during spontaneous and mechanical ventilation of dry gases in sheep. J Appl Physioll990; 69:1117-22. 22. Wanner A. Circulation of the airway mucosa. J Appl Physiol 1989; 67:917-25. 23. McDonald OM. The ultrastructure and permeability of tracheobronchial vessels in health and disease. Eur Respir J 1990; 3(Suppl 12:572s-85s). 24. Persson CGA, Svensjo E. Airway hyperreactivity and microvascular permeability to large molecules. Eur J Respir Dis 1983; 66(Suppl 131: 183s-214s). 25. Lundberg JM, Brodin E, Hua X, Saria A. Vascular permeability changes and smooth muscle contraction in relation to capsaicin-sensitive substance P afferents in the guinea-pig. Acta Physiol Scand 1984; 120:217-27. 26. Saria A, Lundberg JM. Activation of sensory substance P neurones in the respiratory tract by cigarette smoke, mechanical and chemical irritants. Front Hormone Res 1984; 12:123-6. 27. McDonald OM. Respiratory tract infections increase susceptibility to neurogenic inflammation in the rat trachea. Am Rev Respir Dis 1988; 137:1432-40. 28. Persson CGA, Erjefalt I. Terbutaline and adrenaline inhibit leakage of fluid and protein in guinea-pig lung. Eur J Pharmacol1979; 55:199-201. 29. Boschetto P, Roberts NM, Rogers OF, Barnes PJ. Effect of antiasthma drugs on microvascular leakage in guinea-pig airways. Am Rev Respir Dis 1989; 139:416-21. 30. Larrazet F, Chauveau M, Le Gall G, Weber S, Lockhart A, Frossard N. Inhibition of substance P-induced microvascular leakage by inhaled methoxamine in rat airways (abstract). Eur Respir J 1991; 4(Suppl 14:370s-ls). 31. Advenier C, Floch-Saint Aubin A. Bronchopulmonary effects of phenylephrine and methoxamine in the guinea-pig: interaction with bronchoconstrictor drugs. Eur J Pharmacol 1984; 100:59-69. 32. Csete ME, Abraham WM, Wanner A. Vaso-
523
AIRWAY CIRCULATION AND AIRFLOW
motion influences airflow in peripheral airways. Am Rev Respir Dis 1990; 141:1409-13. 33. Hogg JC, Pare PD, Moreno R. The effect of submucosal edema on airways resistance. Am Rev Respir Dis 1987; 135:S54-6. 34. Mariassy AT, Gazeroglu H, Wanner A. Morphometry of the subepithelial circulation in sheep airways: effect of vascular congestion. Am Rev Respir Dis 1991; 143:162-6. 35. Metzger WJ, Zavala D, Richerson HB, et at. Local allergen and local airway inflammation. Am Rev Respir Dis 1987; 135:433-40. 36. Chung KF, KeyesSJ, Morgan BM, Jones PW, Snashall PD. Mechanisms of airway narrowing in acute pulmonary oedema in dogs: influence of the vagus and lung volume. Clin Sci 1983; 65:289-96. 37. Kikuchi R, Sekizawa K, Sasaki H, et at. Effects of pulmonary congestion on airway reactivity to histamine aerosol in dog. J Appl Physiol 1984; 57:1640-7. 38. Kelly L, Kolbe J, Mitzner W, Spannhake EW, Bromberger-Barnea B, Menkes H. Bronchial blood flow affects recovery from constriction in dog lung periphery. J Appl Physiol 1986; 60:1954-9. 39. Csete ME, Chediak AD, Abraham WM, Wanner A. Airway blood flow modifies allergic airway smooth muscle contraction. Am Rev Respir Dis 1991; 144:59-63. 40. Godden DJ, Wagner EM, Pare PD, Mitzner W, Baile EM. Measurement of airway blood flow in sheep by laser-Doppler flowmetry: interpretation and problems. J Appl Physiol 1991; 70:641-9. 41. Hutchroft BJ. Discussion following Hogg JC et at's presentation. In: Lichtenstein LM, Austen KF, eds. Asthma, physiology, immunopharmacology and treatment. New York: Academic Press, 1977; 14-6. 42. Freedman RJ. The functional geometry of the
bronchi: the relationship between changes in external diameter and calibre, and a consideration of the passive role played by the mucosa in bronchoconstriction. Bull Eur Physiopathol Respir 1972; 8:545-52. 43. Inman MD, Watson RM, Killian KJ, O'Byrne PM. Methacholine airway responsiveness decreases during exercisein asthmatic subjects. Am Rev Respir Dis 1990; 141:1414-7. 44. Gilbert lA, Fouke JM, McFadden ER Jr. Heat and water flux in the intrathoracic airways and exercise-induced asthma. J Appl Physiol 1987; 63:1681-91. 45. McFadden ER Jr, Lenner KAM, Strohl KP. Postexertional airway rewarming and thermally induced asthma: new insights into pathophysiology and possible pathogenesis. J Clin Invest 1986; 78:18-25. 46. Broide DH, Eisman S, Ramsdell JW, Ferguson P, Schwartz LB, Wasserman SI. Airway level of mast cell-derived mediators in exercise-induced asthma. Am Rev Respir Dis 1990; 141:563-8. 47. Dinh-Xuan AT, Chaussain M, Regnard J, Lockhart A. Pretreatment with an inhaled a,adrenergic agonist, methoxamine, reduces exerciseinduced asthma. Eur Respir J 1989; 2:409-14. 48. Snashall PD, Boother FA, Sterling GM. The effect of a-adrenergic stimulation on the airways of normal and asthmatic men. Clin Sci 1978; 54:283-9. 49. Black JL, Salome CM, Yan K, Shaw J. Comparison between airways response to an u-adrenoceptor agonist and histamine in asthmatic and nonasthmatic subjects. Br J Clin Pharmacol 1982; 14:464-6. 50. Hardy CC, Bradding P, Robinson C, Holgate ST. Bronchoconstrictor and antibronchoconstrictor properties of inhaled prostacyclin in asthma.
J Appl Physiol 1988; 64:1564-7. 51. Laitinen LA, Laitinen A. Pathology of human asthma. In: Kaliner MA, Barnes PJ, Persson CGA, eds. Asthma: its pathology and treatment. New York: Marcel Dekker, 1991; 103-34. 52. Persson CGA. Tracheobronchial microcirculation in asthma. In: Kaliner MA, Barnes PJ, Persson CGA, eds. Asthma: its pathology and treatment. New York: Marcel Dekker, 1991; 209-30. 53. McNeill RS. Effect of a ~-adrenergic blocking agent, propranolol, on asthmatics. Lancet 1964; 2:1101-2. 54. Snashall PD, Chung KF. Airway obstruction and bronchial hyperresponsiveness in left ventricular failure and mitral stenosis. Am Rev Respir Dis 1991; 144:945-56. 55. Rolla G, Bucca C, Caria E, Scapticci E, Baldi S. Bronchial responsiveness in patients with mitral valve disease. Eur Respir J 1990; 3:127-31. 56. Cabanes LR, Weber SN, Matran R, et at. Bronchial hyperresponsiveness to methacholine in patients with impaired left ventricular function. N Engl J Med 1989; 320:1317-22. 57. Cabanes L, Costes F, WeberS, Regnard J, Benvenuti C, Castaigne A, Guerin F, Lockhart A. Improvement in exercise tolerance by inhalation of methoxamine in patients with impaired left ventricular function. N Engl J Med 1992; 326:1661-5. 58. Regnard J, Baudrillard P, Salah B, Dinh-Xuan AT,Cabanes L, Lockhart A. Inflation of antishock trousers increases bronchial response to methacholine in healthy subjects. J Appl Physiol 1990; 68:1528-33. 59. James AL, Pare PD, Hogg JC. The mechanics of airway narrowing in asthma. Am Rev Respir Dis 1991; 139:242-6.
Introduction As early as 1935, De Burgh Daly (1)proposed that swelling of the bronchial mucosa played a part in the causation of an asthmatic attack and that "the reactions of the bronchomotor ... apparatus of the lungs depend upon the integrity of the bronchial vascular system." In 1943, through use of rigid bronchoscopes, obstructive congestion and edema of intrapulmonary airways were documented in patients with left heart failure (2), and in 1950, they weredocumented in asthmatic subjects after inhaled allergen challenge (3). At about the same time, results of physiologic measurements before and after the subcutaneous administration of epinephrine indicated "... the likelihood that adrenaline produces pulmonary vascular constriction rather than bronchiolar dilatation in certain cases" (4). In 1960, dilatation of the capillary blood vessels beneath the basement membrane and associated mucosal edema were described as "a most striking feature" of the bronchi of subjects who had died in status asthmaticus (5). In 1961, the role of airway narrowing caused by congestion and edema in occurrence of exertional dyspnea and cardiac asthma in congestiveheart failure was again emphasized (6). Thus, the case for v'ascular factors being major contributors to airway obstruction in both left heart failure and asthma was already very strong in the late fifties and early sixties. It is, therefore, surprising that it fellinto oblivion for about 20 yr and that research interest in this area was rekindled only during the last decade. This review will purposely focus on subglottic airways since effects of nasal congestion and vasoconstriction, respectively causing obstruction and increased patency of nasal airways through effects on blood content of venous sinusoids, are fully documented (7). Because there is no available technique to measure tracheobronchial blood flow and its distribution in humans, direct evidencethat airway blood flow influences airflow has not been obtained in humans and was only accrued from studies in experimental animals. In the first part of this report, we review recent experimental work in which the investigators provided simultaneous measures of airway blood flow and lung mechanics. Next, we examine possible mechanisms whereby changes in local blood flow might affect lung mechanics. Lastly, we review human studies in which the investigators attempted to provide evidence of the influence of airway blood flow on airway patency. Studies in Animals Concomitant Bronchial Obstruction and Vasodilatation In intact sheep anesthetized with pentobarbital sodium, inhaled histamine caused an in-
SUMMARY Resistance to gas flow of an airway is a function of both airway smooth muscle tone and thickness of the airway wall internal to the outer ring of airway smooth muscle. Schematically, the increase in airway resistance caused by shortening of airway smooth muscle may be potentiated by a concomitant increase in airway wall thickness caused by vasodilatation of the bronchial vessels and/or microvascular leakage. Conversely, bronchial vasoconstriction may limit to some extent the increase In resistance to gas flow caused by airway smooth muscle shortening and/or congestion and edema of the airway wall. Many endogenous paracrine mediators, putatively Involved in asthma and bronchial hyperresponsiveness, have both bronchomotor and vascular effects. The overall effects on resistance to airflow of endogenous or exogenous agents depend not only upon pre-existing airway smooth muscle tone and pre-existing condition of bronchial vessels but also upon two factors that facilitate microvascular leakage, namely, inflammation of the airway wall and outflow pressure of the bronchial circulation, which is close to left atrial pressure. AM REV RESPIR DIS 1992; 146:819-S23
creasein both bronchial blood flow (measured by an electromagnetic flow probe) and lung resistance with a similar time course, reaching a maximal effect within a fewminutes and progressivelyreturning toward baseline values within 1 h. Bronchial obstruction" and vasodilatation of the bronchial circulation were prevented by the histamine H, antagonist, chlorpheniramine, and the histamine Hz antagonist, metiamide, respectively (8). In pigs anesthetized with pentobarbital, bronchial blood flow was measured with an ultrasonic flow probe while lung resistance and compliance were also recorded. Histamine aerosol caused both bronchial obstruction that was almost completely blocked by the histamine H, antagonist, terfenadine, and vasodilatation that was reduced by only 55 to 60070 by the combination of terfenadine and the histamine Hz antagonist, cimetidine (9). Vagal stimulation also caused both bronchial obstruction and vasodilatation (10). In the same preparation, acetylcholine, nebulized PAF-acether, and prostaglandin D, induced both bronchial obstruction and vasodilatation (9). Vagal-stimulation-induced bronchial obstruction was prevented by atropine, whereas the vasodilatation was resistant to atropine as it was likely due to activation of capsaicinsensitive sensory nerve endings (10). In both sheep and pigs sensitized with Ascaris suum, the early response to nebulized antigens was characterized by a marked and progressive increase in both lung resistance and bronchial blood flow, which peaked within about 10 min and subsided after 90 min. These results were best explained by a histamine-H, -mediated bronchoconstriction and by vasodilatation induced by histamine and cyclooxygenase products. The latter was probably due at least partially to activation of capsaicin-sensitive sensory nerves (9). It was, therefore, hypothesized that changes in bronchial blood flow during the acute response are a sensitive index of mediator re-
lease (11). About half of the sheep had both early and late responses as assessedby changes in lung resistance. Interestingly, the decrease in bronchial vascular resistance preceded the late-phase bronchial obstruction by 60 to 90 min (12). In those experiments, vasodilatation and bronchial obstruction are not necessarily causally linked since both may result from release of mediators by inflammatory cells invading the bronchial wall (12).
Dissociation of Bronchial Obstruction and Vasodilatation Different stimuli caused changes in the bronchial circulation ofpentobarbital-anesthetized pigs with no concomitant changes in lung resistance and dynamic compliance (10). First, inhaled capsaicin in the presence of autonomic blocking agents caused a marked increase in bronchial blood flow but no change in pulmonary mechanics (10). Prior systemic administration of capsaicin significantly reduced capsaicin-induced vasodilatation (10). Furthermore, vasodilatation, which was likely due to release of sensory neuropeptides, was
1 From the Laboratoire de Physiologie Respiratoire, Faculte de Medecine Cochin Port-Royal et Laboratoire d'Explorations Fonctionnelles, Hopital Cochin, Paris, France. 2 Supported by Grant 885012 from INSERM, by GRECO GDRS 15from Centre National de la Recherche Scientifique, Grant 2403Rll from Direction de la Recherche et des Etudes Doctorales, and by Laboratoires Synthelabo. 3 Correspondence and requests for reprints should be addressed to Alain Lockhart, M.D., Laboratoire d'Explorations Fonctionnelles, Hopital Cochin, 27 rue du Faubourg Saint Jacques, 75014 Paris, France. 4 "Bronchial obstruction" is purposely used throughout this report in place of "bronchoconstriction" because the former does not imply a given mechanism, whereas the latter suggests airway smooth muscle contraction alone.
519
520
LOCKHART, DINH-XUAN, REGNARD, CABANES, AND MATRAN
a prominent effect of substance P and/or of calcitonin gene-related peptide in the pig (13). Conversely, the major effect of substance P in the guinea pig is bronchoconstriction which probably accounts for both capsaicin- and ether inhalation-induced bronchial obstruction in this species (14). Second, atropine blocked the vagally induced increase in lung resistance despite the presence of an intact vasodilator response. Similarly, cigarette smoke, calcitonin generelated peptide, bradykinin, and prostacyclin caused bronchial vasodilatation but no change in bronchomotor tone (9). Lastly, on the one hand, the potent bronchial vasodilators prostaglandins E, and E, not only did not reduce airways patency but instead caused bronchodilation (9). On the other hand, vasoconstrictor agents, e.g., a-adrenergic agonists and neuropeptide Y, which increased bronchial vascular resistance, had little if any bronchomotor effect (10). Tracheobronchial blood flow increases with exercisein ponies (15),and with passive hyperventilation (16) and thermal panting in dogs (17). Although they were not measured, one could reasonably assume that lung mechanics were not affected by the increased blood flow in these experiments.
Interplay of Airway Blood Flow and Airflow Summarizing all these results, one is led to the conclusion that in some studies there is a concomitant increase of hindrance to airflow in subglottic airways and in bronchial blood flow, whereas in other experiments, changes in the bronchial circulation are dissociated from changes of lung mechanics in vivo. One could wonder whether it would be possible to reconcile these discrepant findings. Interpretation of simultaneous changes in bronchial blood flow and airflow caused by agents capable of a direct action on bronchial vessels and bronchial smooth muscle is apparently straightforward. For instance, histamine and acetylcholineare potent vasodilators of the tracheobronchial circulation (18), and they contract airway smooth muscle in vitro. It is not surprising, therefore, that they cause a concomitant rise in blood flow and resistance to airflow. However, the relative contribution of shortening of airway smooth muscle and vasodilatation-induced thickening of the airway wall (18) is unknown in vivo. Another apparently simple situation prevails when vasodilatation is associated with bronchodilatation, for example, when the pharmacologic agents used, e.g., prostaglandins E, and E z , are known to relax both vascular and airway smooth muscle. In such cases, effects on the internal diameter of the airways of vasodilatation-induced thickening of the wall is opposed by the effects of bronchial smooth muscle relaxation. However, simple mechanistic explanations cannot account for all experimental data. Indeed, increased bronchial blood flow can be associated with an almost unchanged hindrance to airflow although the agent used is
capable of relaxing (e.g., prostacyclin) or contracting (e.g., tachykinins) airway smooth muscle in vitro. Indeed, several confounding factors may modify the apparent interplay of changes of bronchial blood flow and airflow (figure 1). First, increase in wall thickness of pumpperfused dog trachea varies in an unpredictable manner according to the locally infused vasodilator used, i.e., there was no obvious relation between wall thickness and degree of vasodilatation or known capacity of the agent used for causing microvascular leakage (18). Second, measures of bronchial blood flow with a probe placed around the main bronchial artery tells nothing about blood flow distribution both longitudinally and across the airway wall (19).In the larger bronchi, the subepithelial vascular plexus is connected in parallel to the adventitial plexus by perforating vessels and is drained by bronchial veins. More peripherally, the two plexi merge in a single microvascular network that anastomoses to the alveolar capillaries and drains into pulmonary veins and left atrium (20). In unanesthetized sheep, histamine caused a twofold increase in total systemic blood flow to the airways (19). However, blood flow to the trachea and lobar bronchi increased three and six times over baseline, respectively. Blood flow selectively increased in the mucosal and submucosal layers. The histamine-induced increase in blood flow was to to 15 times over
AirwaywaJJ ttuckness
Endothelium nqhtness Pmv . Pores - Inftarnmation
Clearance from airways walls
Migration of blood cells Activation of mediators eg kinins
Airway smooth muscle tone - direct reflex
Fig. 1. A nonproportional Venn diagram of the interplay of airway smooth muscle tone (bottom circle), tracheobronchial blood flow (Otb) (left upper circle), and endothelium tightness (right upper circle) in the setting of subglotlic airway resistance (Raw). Airway wall thickness is influenced by both vasomotor changes of the tracheobronchial circulation and endothelium tightness. The latter also depends on microvascular pressure and structural changes affecting endothelial cells such as formation of pores caused by pharmacologic agents (e.g., histamine) or local inflammation. Increase in endothelial permeability is associated with diapedesis of inflammatory cells and plasma exudation, which in turn lead to activation of inflammatory peptides (e.g., kinins) and complement moieties. In addition, tracheobronchial blood flow affects clearance of vasoactive and branchoactive substances from the airway wall, thereby modifying magnitude and duration of their effects on the bronchi. Reflexly or directly caused changes in airway smooth muscle tone are major determinants of Raw. Effects on airway caliber of changes in length of airway smooth muscle depend on both site of affected airways and preexisting or concomitant wall thickness. Details in text.
baseline in more peripheral airways (1 to 5 mm in diameter) and five to 15 times over baseline in smaller airways « I mm in diameter). The increase in blood flow was accompanied by wall edema as evidenced by the increase in water content. In these experiments, there was little change in pulmonary blood flow and almost no interstitial or alveolar edema (19).In anesthetized sheep, isocapnic ventilation of dry gas increased blood flow to-fold in the proximal but not in the distal trachea even though mucosal cooling also occurred distally (21). Indeed, factors controlling mucosal blood flow are poorly understood both in large and even more in peripheral airways. In the sheep, tracheal mucosal blood flow lacks cholinergic muscarinic responsiveness. By contrast, tracheal mucosal blood flow shows vasoconstrictor and vasodilator responses to a- and ~-adrenergic agonists that are qualitatively similar to those for total tracheal blood flow (22). Third, effects of a given pharmacologic agent on airway microvascular permeability and, therefore, on airway wall thickness, vary with local conditions. Normal bronchial vessels are not permeable to particulate tracers (23,24). Inflammatory mediators, e.g., histamine, serotonin, sulfidopeptide leukotrienes, and PAF-acether, increase the permeability of the microcirculation in the airways. The increase in permeability is restricted to postcapillary venules and results from active changes in the configuration of the endothelial cells and the opening of gaps between these cells (23). Similarly, both antidromic stimulation of the vagal nerves in atropinized animals, and irritants that stimulate sensory nerve endings cause microvascu1arleakage from postcapillary venulesthrough release of substance P (23-25). Neurogenic microvascular leakage in the airways of experimental animals can be potentiated by mediators, e.g., prostaglandin E" which alone do not increase microvascular permeability in the trachea (26). Naturally occurring respiratory tract infections also increase neurogenic microvascular leakage in the rat airways, as compared with those of specific pathogen-free animals (27). Conversely, administration of a vasoconstrictor agent reduces pharmacologically induced plasma extravasation in the airways (28-30), as well as serotonin-, histamine-, and acetylcholineinduced bronchial obstruction in guinea pigs (31)and nitroglycerin-induced increase in peripheral airway resistance (32). This preventive action is likely due to the reduced microvascular pressure caused by the increase in arteriolar resistance and decreased blood flow in the tracheobronchial circulation. Fourth, congestion, and even more, exudation, caused by either an increase in intravascular pressure (32) or a pharmacologic vasodilatation (18), are capable of increasing airway wall thickness from the trachea down to the small peripheral airways. The effects of wall thickening depend not only on the magnitude of thickening per se but also on the size and location of the affected airways.
521
AIRWAY CIRCULATION AND AIRFLOW
In the trachea and proximal bronchi with a large cross section in relation to the mucosa, it is commonly accepted that mucosal thickening is unlikely to significantly narrow the airway lumen (33,34). However, this contention should be questioned. First, although it certainly applies to large animals, e.g., dogs and sheep, this probably does not hold true for smaller laboratory mammals whose main bronchi range from 1to 2 mm. In smaller cartilaginous and membranous bronchi, the microvascular network is located within the wall, which is made of circular smooth muscle embedded in connective tissue. These are the bronchi where airway wall thickening is most likely to encroach upon the lumen (31), the more so when shortening of airway smooth muscle reduces the external diameter. Indeed, it has been shown that congestion of bronchial vesselsincreased peripheral airway resistanceto gas flow (33).In the smaller bronchioles, which have thinner walls and are surrounded by alveoli, the increased wall thickness occurs at the expense of surrounding air spaces, with little reduction in airway lumen (34). Second, where marked edema is associated with vasodilatation, obstruction of major and second-order bronchi (3) and of sub segmental airways (35) has been documented at bronchoscopy at the time of the early response to allergen challenge in humans. With respect to effects of airway vascular engorgement caused by fluid infusion in animals, it should be emphasized that the concomitant increase ill airway resistance to gas flow results at least partially from reflexly increased vagal tone (36, 37). Therefore, the actual relationship of airway blood flow to airway diameter may well be modified by the pentobarbital anesthesia used in most studies. Lastly, changes in airway blood flow may modify responses to bronchial challenges in vivo..In dogs, the time constant of recovery from histamine-induced increase in peripheral airflow resistance was prolonged with interruption of bronchial blood flow (38). In sensitizedsheep, vasopressinprevented the rise in tracheal blood flow and prolonged the decrease in tracheal volume caused by inhaled Ascaris suum. Nitroglycerin infusion increased tracheal blood flow, and the subsequent administration of Ascaris suum caused neither a further increase in tracheal blood flow nor contraction of the trachea (39).Thus, vasoconstriction likely potentiates the effects of both exogenous and locally released bronchoconstrictor substances through reduction of their vascular clearance from the airway wall. Conversely, increased blood flow may speed up the elimination of spasmogenic substances from the airways, thereby blunting their bronchoconstrictor effects in vivo. In summary, effects of changes in blood flow to the airways on resistance to airflow are not accounted for by changes in airway wall thickness alone. Longitudinal and crosssectional distribution of blood flow, activation of sensory nerve endings, structural changes in the airway wall, modified clearance of bronchoactive substances are con-
founding factors superimposed on the geometric factor. Their effects on airflow may be either reinforced or blunted in experimental animals, much depending on the species studied and the underlying mechanism that predominates. Results of human studies are even more difficult to interpret since no direct measurements of airway blood flow are available. Airway Blood Flow and Airflow in Humans Evidence for effects of changes in blood flow on airway patency is at best circumstantial in humans since none of the available techniques measuring airway blood flow can satisfy both the criteria of accuracy and that of being noninvasive. It was hoped that laserDoppler flowmetry during fiberoptic bronchoscopy might provide direct information on mucosal blood flow in humans. Unfortunately, experiments in live and in dead sheep suggest that in its present form laser-Doppler flowmetry cannot be relied upon to measure airway wall blood flow during bronchoscopy because of background noise at zero flow and large site-to-site variability (40).
chial obstruction in response to u-agonists (48,49), we suggested that the protective effect of methoxamine was likely due to the preventive effects of methoxamine-induced hyperemia and microvascular leakage (47) as opposed to its disproportionate contractile effect on airway smooth musclein unchallenged asthmatics. Indeed, the bronchial response to an agonist may depend on the balance between its vascular effects and its effects on airwaysmooth muscle.Thus, prostacyclinmay cause either bronchodilation or bronchial obstruction in asthmatic subjects (50). Vasodilatation accompanied by plasma extravasation likely occurs spontaneously and during bronchial challenges other than exerciseor hyperventilation in asthmatic subjects. First, inflammatory lesions are found in biopsy specimens of bronchi from asthmatic subjects between asthmatic attacks and even more in symptomatic asthma (51), and vasodilatation and microvascular leakage are common features of inflammation (52). Second, many mediators putatively involvedin asthma and many substances used to demonstrate bronchial hyperresponsiveness in asthmatic subjects, e.g., PAF-acether, histamine, bradykinin, tachykinins, sulfidopeptide leukotrienes, and acetylcholine, with the exception of a-adrenergic agonists and f}-blockers (53, 54), cause vasodilatation and/or plasma exudation. It is likely their effects on airways microvasculature are potentiated in asthmatic subjects because of the presence of inflammatory changes in the airways, as has been demonstrated in the rat (23).
Studies in Asthma Local instillation of antigen under direct vision results in immediate blanching of the mucosa followed within minutes by mucosal edema. The latter is accompanied by hyperemia, so that the mucosa becomes increasingly red, reflecting the congested blood vesselsimmediately beneath the surface of the mucosa (35, 41). Studies in Left Heart Failure Exerciseand hyperventilation of dry and/or cold air are followed by an asthmatic attack Bronchial hyperresponsiveness to the bronin a majority of asthmatic patients. The choconstrictor and vasodilator agents, histapathogenesis of exercise-induced(hyperventi- mine and methacholine, is common in mitral lation-induced) asthma is not fully under- valvedisease and left ventricular failure (55). stood. The thermohydraulic demands on re- Bronchial hyperresponsiveness in such pagional heat and water exchangecertainly result tients is likely due to potentiation of the in an early increase in the tracheobronchial vasodilator properties of histamine and blood flow, almost from the start of exercise methacholine by the high pulmonary and traor hyperventilation. However, bronchial ob- cheobronchial microvascularpressure of these struction is delayed, being maximal only patients. Reasons are as follows. First, there about 5 to 10 min after exercise. There are is a significant correlation between the detwo possible explanations for the delayed gree of bronchial hyperresponsiveness and bronchial response. First, methacholine- pulmonary wedge pressure in patients with induced bronchoconstriction is prevented or mitral valvedisease(55).Second, pretreatment reversed by exerciseand hyperventilation (42, with the inhaled vasoconstrictor agent, 43), which suggests they exert a protective role methoxamine, fully prevented bronchial against nonspecific bronchial hyperreactivity hyperresponsiveness to methacholine in pathrough an unknown mechanism. Second, re- tients with left ventricular failure (56). Third, bound hyperemia probably occurs only af- in those patients, inhaled 132-agonist only ter, and not during, exercise as evidenced by caused partial reversion of methacholinea faster rewarming of the airway wall during induced bronchial obstruction, thus suggestearly recovery in asthmatic compared with ing that the latter could not be attributed to that in normal subjects (44, 45). There is, how- contraction of airway smooth muscle alone ever, no evidence suggesting the release of (56). Clinical implications of bronchial hyperpreformed, or newly formed, inflammatory responsiveness in left ventricular failure and mediators in bronchoalveolar lavage fluid in mitral valve disease are not clear to date. We exercise-induced asthma (46). In addition we have shown that exertional dyspnea is more have shown that pretreatment of inhaled severein patients with left ventricular failure methoxamine, a potent a-adrenergic agonist, and bronchial hyperresponsiveness than in prevents exercise-induced asthma in some those without a response to methacholine asthmatic subjects (47). Because asthmatic (Costes, Cabanes, Lockhart, et al.; personal subjects are exquisitelyprone to develop bron- communication). Furthermore, pretreatment
522
LOCKHART, DINH-XUAN, REGNARD, CABANES, AND MATRAN
Histamine Allergen Bradykinin LTc/D/E4
+
I B2-agonists
i
PGI2
1 1 IRaw\ IIRaw\
Acetylcholine
or 111
1
Ir-R....;..aw-~,....,I
of inhaled methoxamine mar kedly improved exercisecapacity in patients with left ventricular failure in Class III for dyspnea of the New York Heart Association (57). It is therefore likely that exercise-induced vasodilatation facilitates, in the presence of an elevated pulmonary venous (left ventricular enddiastolic) pressure, microvascular leakage in the airway wall, thereby causing some degree of bronchial obstruction in these patients. Consistent with this hypothesis is our finding that inflation of antishock trousers in the erect posture causes a mild degree of bronchial hyperresponsiveness to methacholine in normal subjects (58).
Fig. 2. Airway resistance (Raw) increases when rise in tracheobronchial blood flow (atb) is accompanied by contraction of airway smooth muscle (SM) with (left) or without (right) concomitant increase in permeability (P) of endothelium. Effects of increase in atb on airway resistance when airway smooth muscle tone varies according to balance between vasodilatation and bronchodilatat ion (middle) and to associated changes in microvascular permeability. Same format as figure 1 (t, +, NC = increase, decrease, no change, respectively; LTCIDIE, = leukotrienes C" D" and E,; PGI, = prostacyclin). Details in text.
two situations: (1) when thickening ofthe airway wall is abnormally important because of an elevated microvascular pressure in the airway microcirculation or because of the potentiating effects of inflammatory changes in the airway wall, and (2) when there is concomitant contraction of airway smooth muscle. Morphometric studies of the airways of experimental animals, similar to those carried out to document the role of airway wall thickening in human asthma (59), have not been done to our knowledge. Such studies are probably needed to further our understanding of the effects of changes in blood flow on airflow.
Concluding Remarks
References
Effects of changes in airway blood flow on airflow are far from simple (figures 2 and 3). Indeed, vasodilatation and plasma exudation cause thickening of the airway mucosa. Studies in animals, and circumstantial evidence in humans, suggest that this geometric factor is an important determinant of spontaneous or induced obstruction to airflow in
1. De Burgh Daly I. Interference of intrinsic pulmonary mechanisms as a potential cause of asthma. Edinburgh Med J 1935; 43:139-42. 2. Renault P, Paley PY, Lenegre J, Carouso H. Les alterations bronchiques des cardiaques. J Fr Med Chir Thorac 1943; 3:141-59. 3. Vallery-Radot P, Halpern BN, Dubois de Montreynaud JM, Pean V. Les bronches au cours de la crise d'asthrne. Etude experimentale, bronchoscopique et anatomopathologique. Presse Med 1950; 58:661-4. 4. Shedon MB, Otis AB. Effect of adrenaline on resistance to gas flow in the respiratory tract and on the vital capacity of normal and asthmatic subjects. J Appl Physiol 1951; 9:513-8. 5. Dunnill MS. The pathology of asthma, with special reference to changes in the bronchial mucosa. J Clin Pathol 1960; 13:27-33. 6. Rushmer FJ. Cardiovascular dynamics. 2nd ed. Philadelphia: Saunders, 1961; 470-1. 7. Cole P. Upper respiratory airflow. In: Proctor DF, Andersen 18, eds. The nose: upper airway, physiology and the atmospheric environment. Amsterdam: Elsevier, 1982; 163-90. 8. Long WM, Sprung CL, El Fawal H, et a/. Effects of histamine on bronchial artery blood flow and bronchomotor tone. J Appl Physiol 1985; 59:254-61. 9. Alving K. Airways vasodilatation in the immediate allergic reaction: involvement of inflammatory mediators and sensory nerves. Acta Physiol Scand 1991; 141(Suppl 597:1-64). 10. Matran R. Neural control of lower airway vasculature: involvement of classical transmitters and neuropeptides, Acta Physiol Scand 1991; 142(Suppl 601:1-54). 11. Long WM, Yerger LD, Martinez H, et al. Modification of bronchial blood flow during allergic airway responses. J Appl Physiol 1988;
Alpha-agonists in asthma
Alpha-agonists in LV failure
Fig. 3. A diminished tracheobronchial blood flow can be associated with an increased or decreased airway resistance depending upon magnitude of concomitant changes in both airway smooth muscle (SM) tone and endothelial permeability (P). Inhaled alpha-adrenergic (left) agonists cause moderate to marked bronchial obstruction in most cases of asthma, but they reduce it in the remaining cases, especially when bronchial edema is predominant. These agents invariably reduce methacholine-induced bronchial obstruction in patients with left ventricular failure (right). Details in text. Same format and abbreviations as figures 1 and 2.
65:272-82. 12. Long WM, Yerger LD, Abraham WM, Lobel C. Late-phase bronchial vascular responses in allergic sheep. J Appl Physiol 1990; 69:584-90. 13. Alving K, Matran R, Lacroix JS, Lundberg JM. Capsaicin and histamine antagonist-sensitive mechanisms in the immediate allergic reaction of pig airways. Acta Physiol Scand 1990; 138:49-60. 14. Martling CR, Matran R, Alving K, Hokfelt T, Lundberg JM. Innervation of lower airways and neuropeptide effects on bronchial and vascular tone in the pig. Cell Tissue Res 1990; 260:223-33. 15. Manohar M. Tracheobronchial perfusion during exercise in ponies. J Appl Physiol 1990; 68:2182-5. 16. Baile EM, Godden OJ, Pare PD. Mechanism for increase in tracheobronchial blood flow induced by hyperventilation of dry air in dogs. J Appl Physiol 1990; 68:105-12. 17. Baile EM, Guillemi S, Pare PD. Tracheobronchial and upper airway blood flow in dogs during thermally induced panting. J Appl Physiol 1987; 63:2240-6. 18. Laitinen LA, Robinson NP, Laitinen A, Widdicombe J G. Relationship between tracheal mucosal thickness and vascular resistance in dogs. J Appl Physiol 1986; 61:2186-93. 19. Kramer GG, Lindsey C, Wu CH, Mertens S, Russell LA, Cross CEo Airway blood flow distribution and lung edema after histamine infusion in awake sheep. J Appl Physiol1988; 65:1847-54. 20. Deffenbach ME, Charan NB, Lakshminarayan S, Butler J. The bronchial circulation: small, but a vital attribute of the lung. Am Rev Respir Dis 1987; 135:463-81. 21. White DA, Parsons GH. Tracheal blood flow during spontaneous and mechanical ventilation of dry gases in sheep. J Appl Physioll990; 69:1117-22. 22. Wanner A. Circulation of the airway mucosa. J Appl Physiol 1989; 67:917-25. 23. McDonald OM. The ultrastructure and permeability of tracheobronchial vessels in health and disease. Eur Respir J 1990; 3(Suppl 12:572s-85s). 24. Persson CGA, Svensjo E. Airway hyperreactivity and microvascular permeability to large molecules. Eur J Respir Dis 1983; 66(Suppl 131: 183s-214s). 25. Lundberg JM, Brodin E, Hua X, Saria A. Vascular permeability changes and smooth muscle contraction in relation to capsaicin-sensitive substance P afferents in the guinea-pig. Acta Physiol Scand 1984; 120:217-27. 26. Saria A, Lundberg JM. Activation of sensory substance P neurones in the respiratory tract by cigarette smoke, mechanical and chemical irritants. Front Hormone Res 1984; 12:123-6. 27. McDonald OM. Respiratory tract infections increase susceptibility to neurogenic inflammation in the rat trachea. Am Rev Respir Dis 1988; 137:1432-40. 28. Persson CGA, Erjefalt I. Terbutaline and adrenaline inhibit leakage of fluid and protein in guinea-pig lung. Eur J Pharmacol1979; 55:199-201. 29. Boschetto P, Roberts NM, Rogers OF, Barnes PJ. Effect of antiasthma drugs on microvascular leakage in guinea-pig airways. Am Rev Respir Dis 1989; 139:416-21. 30. Larrazet F, Chauveau M, Le Gall G, Weber S, Lockhart A, Frossard N. Inhibition of substance P-induced microvascular leakage by inhaled methoxamine in rat airways (abstract). Eur Respir J 1991; 4(Suppl 14:370s-ls). 31. Advenier C, Floch-Saint Aubin A. Bronchopulmonary effects of phenylephrine and methoxamine in the guinea-pig: interaction with bronchoconstrictor drugs. Eur J Pharmacol 1984; 100:59-69. 32. Csete ME, Abraham WM, Wanner A. Vaso-
523
AIRWAY CIRCULATION AND AIRFLOW
motion influences airflow in peripheral airways. Am Rev Respir Dis 1990; 141:1409-13. 33. Hogg JC, Pare PD, Moreno R. The effect of submucosal edema on airways resistance. Am Rev Respir Dis 1987; 135:S54-6. 34. Mariassy AT, Gazeroglu H, Wanner A. Morphometry of the subepithelial circulation in sheep airways: effect of vascular congestion. Am Rev Respir Dis 1991; 143:162-6. 35. Metzger WJ, Zavala D, Richerson HB, et at. Local allergen and local airway inflammation. Am Rev Respir Dis 1987; 135:433-40. 36. Chung KF, KeyesSJ, Morgan BM, Jones PW, Snashall PD. Mechanisms of airway narrowing in acute pulmonary oedema in dogs: influence of the vagus and lung volume. Clin Sci 1983; 65:289-96. 37. Kikuchi R, Sekizawa K, Sasaki H, et at. Effects of pulmonary congestion on airway reactivity to histamine aerosol in dog. J Appl Physiol 1984; 57:1640-7. 38. Kelly L, Kolbe J, Mitzner W, Spannhake EW, Bromberger-Barnea B, Menkes H. Bronchial blood flow affects recovery from constriction in dog lung periphery. J Appl Physiol 1986; 60:1954-9. 39. Csete ME, Chediak AD, Abraham WM, Wanner A. Airway blood flow modifies allergic airway smooth muscle contraction. Am Rev Respir Dis 1991; 144:59-63. 40. Godden DJ, Wagner EM, Pare PD, Mitzner W, Baile EM. Measurement of airway blood flow in sheep by laser-Doppler flowmetry: interpretation and problems. J Appl Physiol 1991; 70:641-9. 41. Hutchroft BJ. Discussion following Hogg JC et at's presentation. In: Lichtenstein LM, Austen KF, eds. Asthma, physiology, immunopharmacology and treatment. New York: Academic Press, 1977; 14-6. 42. Freedman RJ. The functional geometry of the
bronchi: the relationship between changes in external diameter and calibre, and a consideration of the passive role played by the mucosa in bronchoconstriction. Bull Eur Physiopathol Respir 1972; 8:545-52. 43. Inman MD, Watson RM, Killian KJ, O'Byrne PM. Methacholine airway responsiveness decreases during exercisein asthmatic subjects. Am Rev Respir Dis 1990; 141:1414-7. 44. Gilbert lA, Fouke JM, McFadden ER Jr. Heat and water flux in the intrathoracic airways and exercise-induced asthma. J Appl Physiol 1987; 63:1681-91. 45. McFadden ER Jr, Lenner KAM, Strohl KP. Postexertional airway rewarming and thermally induced asthma: new insights into pathophysiology and possible pathogenesis. J Clin Invest 1986; 78:18-25. 46. Broide DH, Eisman S, Ramsdell JW, Ferguson P, Schwartz LB, Wasserman SI. Airway level of mast cell-derived mediators in exercise-induced asthma. Am Rev Respir Dis 1990; 141:563-8. 47. Dinh-Xuan AT, Chaussain M, Regnard J, Lockhart A. Pretreatment with an inhaled a,adrenergic agonist, methoxamine, reduces exerciseinduced asthma. Eur Respir J 1989; 2:409-14. 48. Snashall PD, Boother FA, Sterling GM. The effect of a-adrenergic stimulation on the airways of normal and asthmatic men. Clin Sci 1978; 54:283-9. 49. Black JL, Salome CM, Yan K, Shaw J. Comparison between airways response to an u-adrenoceptor agonist and histamine in asthmatic and nonasthmatic subjects. Br J Clin Pharmacol 1982; 14:464-6. 50. Hardy CC, Bradding P, Robinson C, Holgate ST. Bronchoconstrictor and antibronchoconstrictor properties of inhaled prostacyclin in asthma.
J Appl Physiol 1988; 64:1564-7. 51. Laitinen LA, Laitinen A. Pathology of human asthma. In: Kaliner MA, Barnes PJ, Persson CGA, eds. Asthma: its pathology and treatment. New York: Marcel Dekker, 1991; 103-34. 52. Persson CGA. Tracheobronchial microcirculation in asthma. In: Kaliner MA, Barnes PJ, Persson CGA, eds. Asthma: its pathology and treatment. New York: Marcel Dekker, 1991; 209-30. 53. McNeill RS. Effect of a ~-adrenergic blocking agent, propranolol, on asthmatics. Lancet 1964; 2:1101-2. 54. Snashall PD, Chung KF. Airway obstruction and bronchial hyperresponsiveness in left ventricular failure and mitral stenosis. Am Rev Respir Dis 1991; 144:945-56. 55. Rolla G, Bucca C, Caria E, Scapticci E, Baldi S. Bronchial responsiveness in patients with mitral valve disease. Eur Respir J 1990; 3:127-31. 56. Cabanes LR, Weber SN, Matran R, et at. Bronchial hyperresponsiveness to methacholine in patients with impaired left ventricular function. N Engl J Med 1989; 320:1317-22. 57. Cabanes L, Costes F, WeberS, Regnard J, Benvenuti C, Castaigne A, Guerin F, Lockhart A. Improvement in exercise tolerance by inhalation of methoxamine in patients with impaired left ventricular function. N Engl J Med 1992; 326:1661-5. 58. Regnard J, Baudrillard P, Salah B, Dinh-Xuan AT,Cabanes L, Lockhart A. Inflation of antishock trousers increases bronchial response to methacholine in healthy subjects. J Appl Physiol 1990; 68:1528-33. 59. James AL, Pare PD, Hogg JC. The mechanics of airway narrowing in asthma. Am Rev Respir Dis 1991; 139:242-6.
